Prediction of natural gas flow through chokes using support vector machine algorithm

摘要: Abstract In oil and gas fields, it is a common practice to flow liquid mixtures through choke valves. general, different types of primary valves are employed control pressure rate when the producing well directs natural processing equipment. this case, valve normally affected by elevated levels (or velocity) as solid materials suspended in phase (e.g., fine sand other debris). Both surface subsurface chokes may be installed regulate rates protect porous medium facilities from unusual instabilities. study reliable, novel, computer based predictive model using Least-Squares Support Vector Machine (LSSVM) algorithm applied predict coefficient both nozzle orifice type subsonic conditions. The average absolute relative deviation proposed reported data for nozzle-type orifice-type nearly 0.25% 0.15% squared correlation around 0.9961 0.9982 respectively.